Netflix is starting 2025 with a new mystery series from Harlan Coben called Missing You, which now has an official trailer.Detective Kat Donovan (Rosalind Eleazar) has found the man of his dreams in Josh Buchanan (Ashley Walters). The couple get engaged, and things look like theyre heading toward a perfect ending until tragedy strikes. Josh disappears, leaving no clues as to where he went. Eleven years later, Kat is trying to move on and does so by signing up for a dating app. In a cruel twist of fate, Kat finds Joshs profile, which triggers her grief. Per Netflix, Joshs reappearance forces her to dive back into the mystery surrounding her fathers murder and uncover long-buried secrets from her past.Recommended VideosMissing Yousensemble includesJessica PlummerMISSING YOU | Official Trailer | NetflixMissing Youis based on Cobens 2014 novel of the same name. Though the novel takes place in New York City, the Netflix series relocates the setting to the U.K. Victoria Asare-Archer is the lead writer, with Nimer Rashed and Isher Sahota directing. Executive producers include Coben, Nicola Shindler, Asare-Archer, Richard Fee, and Danny Brocklehurst.RelatedCobens novels have been the source material for nearly a dozen Netflix series, includingThe Stranger,Stay Close, and Harlan Cobens Shelter. Cobens last series, Fool Me Once, generated record-breaking viewership after its premiere on January 1, 2024. Fool Me Onceis Netflixs eighth-most popular English-language TV series.Missing Youpremieres on January 1, 2025.Editors Recommendations

Dana said CEO and Chairman James Kamsickas retired after nearly a decade and it named Bruce McDonald his successor while it initiates a cost-cutting strategy and explores the sale of a unit.

Leaky cyber defenses allowed customer data to be stolen, enabling Covid-19 fraud, attorney general alleges

A bureaucrats promotion comes at a moral cost in this drama directed by Mohammad Rasoulof, who was sentenced to prison during its production and has since fled the country.

This NOVA presentation follows engineers as they seek to help humanity adapt to a changing world, drawing on the ideas and traditions of the past to create new technologies.

You've never heard anything like it Nvidias new AI audio model can synthesize sounds that have never existed What does a screaming saxophone sound like? The Fugatto model has an answer... Kyle Orland Nov 25, 2024 4:40 pm | 39 An audio wave can contain so much. An angry cello, for instance... Credit: Getty Images An audio wave can contain so much. An angry cello, for instance... Credit: Getty Images Story textSizeSmallStandardLargeWidth *StandardWideLinksStandardOrange* Subscribers only Learn moreAt this point, anyone who has been following AI research is long familiar with generative models that can synthesize speech or melodic music from nothing but text prompting. Nvidia's newly revealed "Fugatto" model looks to go a step further, using new synthetic training methods and inference-level combination techniques to "transform any mix of music, voices, and sounds," including the synthesis of sounds that have never existed.While Fugatto isn't available for public testing yet, a sample-filled website showcases how Fugatto can be used to dial a number of distinct audio traits and descriptions up or down, resulting in everything from the sound of saxophones barking to people speaking underwater to ambulance sirens singing in a kind of choir. While the results on display can be a bit hit or miss, the vast array of capabilities on display here helps support Nvidia's description of Fugatto as "a Swiss Army knife for sound."Youre only as good as your dataIn an explanatory research paper, over a dozen Nvidia researchers explain the difficulty in crafting a training dataset that can "reveal meaningful relationships between audio and language." While standard language models can often infer how to handle various instructions from the text-based data itself, it can be hard to generalize descriptions and traits from audio without more explicit guidance.To that end, the researchers start by using an LLM to generate a Python script that can create a large number of template-based and free-form instructions describing different audio "personas" (e.g., "standard, young-crowd, thirty-somethings, professional"). They then generate a set of both absolute (e.g., "synthesize a happy voice") and relative (e.g., "increase the happiness of this voice") instructions that can be applied to those personas.The wide array of open source audio datasets used as the basis for Fugatto generally don't have these kinds of trait measurements embedded in them by default. But the researchers make use of existing audio understanding models to create "synthetic captions" for their training clips based on their prompts, creating natural language descriptions that can automatically quantify traits such as gender, emotion, and speech quality. Audio processing tools are also used to describe and quantify training clips on a more acoustic level (e.g. "fundamental frequency variance" or "reverb").For relational comparisons, the researchers rely on datasets where one factor is held constant while another changes, such as different emotional readings of the same text or different instruments playing the same notes. By comparing these samples across a large enough set of data samples, the model can start to learn what kinds of audio characteristics tend to appear in "happier" speech, for instance, or differentiate the sound of a saxophone and a flute.After running a variety of different open source audio collections through this process, the researchers ended up with a heavily annotated dataset of 20 million separate samples representing at least 50,000 hours of audio. From there, a set of 32 Nvidia tensor cores was used to create a model with 2.5 billion parameters that started to show reliable scores on a variety of audio quality tests.Its all in the mix OK, Fugatto, can we get a little more barking and a little less saxophone in the monitors? Credit: Getty Images Beyond the training, Nvidia is also talking up Fugatto's "ComposableART" system (for "Audio Representation Transformation"). When provided with a prompt in text and/or audio, this system can use "conditional guidance" to "independently control and generate (unseen) combinations of instructions and tasks" and generate "highly customizable audio outputs outside the training distribution." In other words, it can combine different traits from its training set to create entirely new sounds that have never been heard before.I won't pretend to understand all of the complex math described in the paperwhich involves a "weighted combination of vector fields between instructions, frame indices and models." But the end results, as shown in examples on the project's webpage and in an Nvidia trailer, highlight how ComposableART can be used to create the sound of, say, a violin that "sounds like a laughing baby or a banjo that's playing in front of gentle rainfall" or "factory machinery that screams in metallic agony." While some of these examples are more convincing to our ears than others, the fact that Fugatto can take a decent stab at these kinds of combinations at all is a testament to the way the model characterizes and mixes extremely disparate audio data from multiple different open source data sets.Perhaps the most interesting part of Fugatto is the way it treats each individual audio trait as a tunable continuum, rather than a binary. For an example that melds the sound of an acoustic guitar and running water, for instance, the result ends up very different when either the guitar or the water is weighted more heavily in Fugatto's interpolated mix. Nvidia also mentions examples of tuning a French accent to be heavier or lighter, or varying the "degree of sorrow" inherent in a spoken clip.Beyond tuning and combining different audio traits, Fugatto can also perform the kinds of audio tasks we've seen in previous models, like changing the emotion in a piece of spoken text or isolating the vocal track in a piece of music. Fugatto can also detect individual notes in a piece of MIDI music and replace them with a variety of vocal performances, or detect the beat of a piece of music and add effects from drums to barking dogs to ticking clocks in a way that matches the rhythm. Fugatto's generated audio (magenta) matches the melody of an input MIDI file (Cyan) very closely. Credit: Nvidia Research While the researchers describe Fugatto as just the first step "towards a future where unsupervised multitask learning emerges from data and model scale," Nvidia is already talking up use cases from song prototyping to dynamically changing video game scores to international ad targeting. But Nvidia was also quick to highlight that models like Fugatto are best seen as a new tool for audio artists rather than a replacement for their creative talents."The history of music is also a history of technology," Nvidia Inception participant and producer/songwriter Ido Zmishlany said in Nvidia's blog post. "The electric guitar gave the world rock and roll. When the sampler showed up, hip-hop was born. With AI, were writing the next chapter of music. We have a new instrument, a new tool for making musicand thats super exciting."Kyle OrlandSenior Gaming EditorKyle OrlandSenior Gaming Editor Kyle Orland has been the Senior Gaming Editor at Ars Technica since 2012, writing primarily about the business, tech, and culture behind video games. He has journalism and computer science degrees from University of Maryland. He once wrote a whole book about Minesweeper. 39 Comments Prev story

Vicarious infringement Supreme Court wants US input on whether ISPs should be liable for users piracy SCOTUS asks US government for its view on $1 billion Sony v. Cox case. Jon Brodkin Nov 25, 2024 4:15 pm | 57 Credit: Getty Images | OcusFocus Credit: Getty Images | OcusFocus Story textSizeSmallStandardLargeWidth *StandardWideLinksStandardOrange* Subscribers only Learn moreThe Supreme Court signaled it may take up a case that could determine whether Internet service providers must terminate users who are accused of copyright infringement. In an order issued today, the court invited the Department of Justice's solicitor general to file a brief "expressing the views of the United States."In Sony Music Entertainment v. Cox Communications, the major record labels argue that cable provider Cox should be held liable for failing to terminate users who were repeatedly flagged for infringement based on their IP addresses being connected to torrent downloads. There was a mixed ruling at the US Court of Appeals for the 4th Circuit as the appeals court affirmed a jury's finding that Cox was guilty of willful contributory infringement but reversed a verdict on vicarious infringement "because Cox did not profit from its subscribers' acts of infringement."That ruling vacated a $1 billion damages award and ordered a new damages trial. Cox and Sony are both seeking a Supreme Court review. Cox wants to overturn the finding of willful contributory infringement, while Sony wants to reinstate the $1 billion verdict.The Supreme Court asking for US input on Sony v. Coxcould be a precursor to the high court taking up the case. For example, the court last year asked the solicitor general to weigh in on Texas and Florida laws that restricted how social media companies can moderate their platforms. The court subsequently took up the case and vacated lower-court rulings, making it clear that content moderation is protected by the First Amendment.Record labels vs. ISPsCox has said that letting the piracy ruling stand "would force ISPs to terminate Internet service to households or businesses based on unproven allegations of infringing activity, and put them in a position of having to police their networks." Cox said that ISPs "have no way of verifying whether a bot-generated notice is accurate" and that even if the notices are accurate, terminating an account would punish every user in a household where only one person may have illegally downloaded copyrighted files.Record labels urged the court to reinstate the vicarious infringement verdict. "As the District Court explained, the jury had ample evidence that Cox profited from its subscribers' infringement, including evidence 'that when deciding whether to terminate a subscriber for repeat infringement, Cox considered the subscriber's monthly payments,' and 'Cox repeatedly declined to terminate infringing subscribers' Internet service in order to continue collecting their monthly fees,'" the record labels' petition said.Another potentially important copyright case involves the record labels and Grande, an ISP owned by Astound Broadband. The conservative-leaning US Court of Appeals for the 5th Circuit ruled last month that Grande violated the law by failing to terminate subscribers accused of being repeat infringers. The 5th Circuit also ordered a new trial on damages because it said a $46.8 million award was too high. Grande and the record labels are both seeking en banc rehearings of the 5th Circuit panel ruling.Jon BrodkinSenior IT ReporterJon BrodkinSenior IT Reporter Jon is a Senior IT Reporter for Ars Technica. He covers the telecom industry, Federal Communications Commission rulemakings, broadband consumer affairs, court cases, and government regulation of the tech industry. 57 Comments

Max Belov, Chief Technology Officer, Coherent SolutionsNovember 25, 20244 Min ReadMopic via Alamy StockEvery breakthrough has its share of mistakes. Artificial intelligence is disrupting routine tasks and is quickly establishing itself as a very powerful personal assistant. For example, AI helps medical researchers find and evaluate available donors for cell treatments, giving patients hope where there was none -- and the list of AI uses goes on. Yet, this same technology generates misleading financial forecasts based on non-existent data or creates references to fictitious scientific articles.AI models are only as trustworthy as the data they are trained on. However, even with a solid data foundation, the results of AI predictions are not 100% accurate. The impact of their occasional hallucinations may range from causing slight user embarrassment to billions of dollars worth of financial losses and legal repercussions for organizations. The question is how organizations can look beyond hallucinations and rely on AI in decision-making when the models are partially transparent.AI Confidence Mislead Decision-MakersOver half of Fortune 500 companies note AI as a potential risk factor. They fear AI inconsistencies and potential ethical risks that might lead to negative brand publicity and financial and reputational losses.It is impossible to fix AI hallucinations with a wave of a hand. So far, hallucinations are a common challenge in AI solutions. While the explainability of traditional ML methods and neural networks is well understood by now, many researchers are working on methods to explain GenAI and LLMs. Significant advancements will come in the near future. Meanwhile, AI certainly shouldnt be dismissed because it's not entirely reliable: It has already become a must-have tool for organizations across various industries. Decision-makers should rely on human intelligence and supervision to effectively integrate AI models into business processes.Related:Black Box Trust IssuesAI models are black boxes that lack transparency and are only partially explainable. Hallucinations are common in complex language models and deep learning systems. Such systems are affected since they hinge on patterns derived from vast datasets rather than on a fundamental deterministic understanding of the content.The good news is that taking an insightful look into the black boxes is possible, to a certain extent. Organizations can use specific methods to address one of the major trust issues with AI.Explaining the UnexplainableIn many business applications, especially those influencing critical decision-making, the ability to explain how an AI model reaches its conclusions is more important than achieving the highest possible models accuracy.Related:Not all AI models are black boxes. For example, decision trees or linear regressions are common in predictive analytics, financial forecasting, and business intelligence applications. These types of AI models are interpretable.For non-transparent models, SHAP (shapley additive explanations) helps explain how much each input affects an LLMs prediction. For example, users can ask an LLM to highlight key points in the input data and explain the logical chain behind the output. The answers can help improve system prompts and input data. However, SHAP has limited effectiveness for pre-trained LLMs due to their complexity, which requires different methods to explain their results. This is still a very rapidly developing field, with new emerging approaches for the interpretability of LLMs, such as using attention mechanisms to trace back how a model reaches its conclusion or using LLMs with memory functions to reduce inconsistencies over time.How Can Organizations Rely on AI Models?Organizations should carefully manage and contextualize the reliability of the models they use. Decision-makers can apply guardrails like regular audits and protocols for human oversight. They can consider creating a domain-specific knowledge base, which, for example, will be paramount for medical professionals, as their decisions often impact people's lives. They can also apply theRAGapproach (retrieval augmented generation) to mitigate associated risks. For example, a customer support chatbot can retrieve past interactions with a client, augment that data with product updates, and generate highly relevant responses to resolve a query.Related:Generative AI works best by augmenting human decision-making rather than entirely replacing it. It is important to keep humans in the loop, as they are competent to monitor a models accuracy and ethical compliance. As a rule of thumb, implement GenAI solutions that provide insights while putting human employees in charge of making the final decisions. They can correct and refine the outputs before an AI-driven error grows into a problem.AI models should be dynamic. Feedback loops where humans report issues and introduce changes play a key role in maintaining and enhancing the accuracy and reliability of AI. The next step in aligning AI with organizational processes is in fostering collaboration between data scientists, domain experts, and leaders.Lastly, before investing in GenAI, organizations should conduct a maturity assessment to make sure they have the necessary data infrastructure and robust governance policies in place. They need these to enhance the quality and accessibility of data used to train AI models.AI has great potential to enhance decision-making, but organizations must acknowledge the risks of hallucinations. When they implement consistent measures addressing this issue, they build trust in AI and maximize the benefits of AI solutions.About the AuthorMax BelovChief Technology Officer, Coherent SolutionsMax Belov joined Coherent Solutions in 1998 and assumed the role of CTO two years later. He is a seasoned software architect with deep expertise in designing and implementing distributed systems, cybersecurity, cloud technology, and AI. He also leads Coherents R&D Lab, focusing on IoT, blockchain, and AI innovations. His commentary and bylines appeared in CIO, Silicon UK Tech News, Business Reporter, and TechRadar Pro.See more from Max BelovNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also LikeReportsMore Reports

Plastic waste in IndonesiaPA Images / AlamyThe world currently produces more than 50 million tonnes of mismanaged plastic waste each year, and some researchers project this flood of plastic pollution will double by mid-century but they also say that, if countries can agree to adopt four key policies during global plastic treaty negotiations this week, we could slash that number by 90 per cent.Plastic pollution ends up clogging ecosystems on land and at sea. This has an impact on every level of the food chain, from phytoplankton cells to humans, says Sarah-Jeanne Royer at the University of California, San Diego. Plastics are also responsible for about 5 per cent of greenhouse gas emissions.Thats why most of the worlds countries are meeting in Busan, South Korea this week to hammer out the final details of a global treaty aimed at ending plastic pollution. In 2022, 175 countries already agreed to adopt the legally binding treaty and have spent the past two years debating exactly what it should require, with particular disagreements over setting limits on the production of new plastic.AdvertisementTo bring more clarity to the debate, Douglas McCauley at the University of California, Santa Barbara and his colleagues used an artificial intelligence model trained on economic data to test how the policies under consideration would affect global plastic pollution. I wasnt convinced that [eliminating plastic pollution] was actually possible, says McCauley. But it turns out you can get pretty darn close.According to their projections, under current conditions, plastic pollution is set to roughly double to between 100 and 139 million tonnes by 2050. But a combination of four policies, all of which are still on the table in the current treaty draft, were enough to reduce this by more than 90 per cent. Get a dose of climate optimism delivered straight to your inbox every month.Sign up to newsletterThe most impactful of these was a mandate that plastic products contain at least 40 per cent recycled material. That rule alone cut plastic pollution in half by mid-century. This effect is so significant because it cuts demand for newly made or virgin plastic while also spurring demand for recycled materials, says McCauley. Suddenly theres a giant global market for recycling.But recycling on its own wasnt sufficient. If your target is to end plastic pollution, you need to do things across the entire lifecycle, he says. Deeper cuts required limiting production of virgin plastics to 2020 levels. This production cap cut plastic pollution by around 60 million tonnes per year by the middle of the century, according to the model. This change also had the greatest impact on greenhouse gas emissions from plastic production, as extracting fossil fuels and turning them into virgin plastics involves emissions-intensive processes.A third policy, spending $50 billion on waste management, reduced pollution by nearly the same amount as the production cap especially if these funds were spent in low-income countries with poor infrastructure, which are also the most inundated by plastic pollution. When you start talking about global finance, [the amount of money needed] is not that big, says McCauley. Building a sanitary landfill is not like building a port.Plastic waste is increasing, and though some is recycled or destroyed, a large portion is mismanaged and piles up as plastic pollutionA. Samuel Pottinger et al.Finally, a small tax on plastic packaging cut pollution by tens of millions of tonnes. The researchers based this estimate on case studies of how people reduced their plastic use in response to similar taxes, such as a 5 cent fee on single-use plastic bags in Washington DC. Money raised by such a tax could also be used to pay for other changes, like building out waste management infrastructure or improving recycling systems.Royer, who was not involved with the study, says she thinks those policies would all help. Reducing the use of single-use plastic such as grocery bags or plastic forks via a tax or a ban could also make a difference, she says. If we look at plastic pollution in general, 40 per cent of the plastic being produced is single-use items.However, she points out local rules alone will never solve the problem. For instance, California banned some single-use plastic bags a decade ago and this year banned all such bags. But most of the plastic pollution that washes up on its beaches originates outside the state: Californias plastic waste generally drifts across the Pacific from Asia or is flotsam left by the fishing industry. Theres no border, says Royer.Thats where a global treaty comes in. The researchers showed how implementing different policies across the world would cut down on three things: the volume of mismanaged plastic waste, the production of new plastics, and plastic-related greenhouse gas emissions. The four key policies in combination, seen in the graph below, reduced all three measures, and in particular slashed mismanaged waste by 91 per cent.Researchers estimated the impact of different policies for reducing plastic wasteA. Samuel Pottinger et al.In Busan, countries have now reached the deadline to decide on a final treaty draft, but they remain far apart on key issues. A main fault line is whether the treaty should include a production cap on newly made plastics, which the researchers found was the second-most impactful policy. Plastic-producing countries and the petrochemical industry oppose production caps, instead throwing their support behind recycling measures.A high-ambition coalition of 68 countries, including the UK, is pushing for a treaty that would include both, with the goal of eliminating plastic pollution by 2040. Other researchers have also argued a cap on plastic production is necessary to end pollution. But just last week, advocates for a production cap were dismayed by reports the US would not support a specific limit on plastic production. McCauley recently penned an open letter signed by more than one hundred researchers to the Biden administration urging it to support a strong plastic treaty.Were at a pivotal moment, said Erin Simon at the World Wildlife Foundation, an environmental advocacy group, in an email to press. Our last best chance to forge an agreement that could end the flow of plastic into nature is within reach, but only if countries come to the negotiating table with a clear vision and determination to get the job done.Journal referenceScience DOI: 10.1126/science.adr3837Topics:

The H5N1 bird flu virus has caused sporadic cases in humansLuca Bruno/Associated Press/AlamyIf the H5N1 bird flu virus sparks a pandemic, older people may have better immunity than younger people because of past exposure to closely related viruses.In previous clusters of H5N1 cases, older people were less likely to catch the virus than younger individuals, so scientists hypothesised that the former group may have stronger immunity against the virus than those who are younger. This could be because older people have had more exposure to flu strains that are closely related to H5N1, called H1N1 and H2N2.